Security Engineering Research Group

Virtualization on Mobiles

“VMware on Monday extended its virtualization platform to the mobile phone platform with the unveiling of its VMware Mobile Virtualization Platform based on a previously unannounced acquisition it closed last month.The new platform lets mobile phone users work with multiple virtual phones at the same time, said Srinivas Krishnamurti, VMware’s director of product development management and market development.”

http://www.crn.com/networking/212001314

SACMAT2009 Call for Papers (CFP)

SACMAT 2009 call for papers is out:

http://www.sacmat.org/2009/date.php

*  January 16
o Deadline for submission of paper
* January 16
o Deadline for submission of panel proposal
* March 6
o Acceptance notification
* March 31
o Deadline for camera-ready copies of the accepted papers
* June 3-5
o SACMAT in Stresa, Italy

Mobilware Call for Papers (CFP)

Mobilware in Berlin, Germany:

http://www.mobilware.org/cfp.shtml

Important Dates

Paper Submission Due Date:     November 23, 2008
Workshop/Tutorial Proposals Due Date:     December 1, 2008
Notification of Acceptance:     February 1, 2009
Camera Ready Versions Due:     February 20, 2009
Conference Dates:     April 28-30, 2009

Backing up OpenMoko FR before bricking it and getting your custom kernel and rootfs onto it …

After successfully configuring Open Embedded http://recluze.wordpress.com/2008/09/07/getting-started-with-openmoko/, the build environment for OpenMoko, we want to try our custom built kernel and rootfs. Although, we will be using Micro SD card to run our custom builds rather than the flash still we will backup the factory default images because you never know when you brick it in stupidity or frustration.

Flash memories are fast but there are two things that need to be kept in mind 1) Flash has a limited age with respect to the number of flashes (maybe a 100,000) 2) And we only flash those images onto the flashes, which are satisfactory for usage and are not experimental. Micro SD has one limitation that it is slower in performance compared to Flash memory. So if you live near an OpenMoko store it’ll be better to use the Flash off and on but if you live in Peshawar than you should prefer to use the Micro SD card.

Backing up the Original Images

(This has been adopted from http://wiki.openmoko.org/wiki/Backup to suit newbies like us)

There are two ways to backup:

1. DFU-Util (The official and buggy way of doing it)
2. Common Sense

DFU-Util

(Adopted from http://wiki.openmoko.org/wiki/Dfu-util)

DF stands for Device Firmware and U stands for utility. DFU-Util is a host implementation of Universal Serial Bus Device Firmware Upgrade protocol. The device side is implemented in the uBoot of OpenMoko.

1. transfer and flash partitions in NAND.
2. transfer anything into RAM
   • this can be used for fast development cycles of low-level code such as kernels without flashing them (Haven’t tried it yet)
3. Read out the current internal NAND
   • this is an easy and efficient way of doing full backups of your phone

If you are not sure what NAND and NOR are the read http://www.linuxdevices.com/articles/AT7478621147.html

Installing DFU-Util
Debian/Ubuntu
sudo apt-get install libusb-dev autogen pkg-config autotools-dev autoconf automake

Fedora Core
yum install libusb-devel libusb-static

Or download a binary for Linux from http://downloads.openmoko.org/releases/Om2008.8-update/defu-util

Or if you have built Open Embedded using the mokoMakefile then you should have the binary in <$MOKODIR>/build/tmp/staging/i686-linux/usr/bin

Using DFU-Util

Now that we are ready ensure that you are using the USB Data Cable (stay away from USB HUBs) and you are logged as root on your host PC. Power off the phone. Now hold the AUX (top left of phone) button and then hold the power button until the NOR boot menu appears.

./dfu-util -a kernel -R -U original-kernel.bin
./dfu-util -a rootfs -R -U original-rootfs.jffs2
./dfu-util -a splash -R -U original-splash.bin
./dfu-util -a u-boot -R -U original-u-boot.bin
./dfu-util -a u-boot_env -R -U original-u-boot_env.bin

If you are successful then you are very lucky We were not as lucky as you were and we had to do it over and over again and still we had a corrupt kernel in our backups.

Using Common Sense

–TODO–

Recluze: I only remember the rootfs part so can you plz complete this….

Flashing kernel and rootfs onto Micro SD Card

(Adopted from http://wiki.openmoko.org/wiki/Booting_from_SD.html)

Preparing the SD card

We got our SD card plugged in to our laptop via USB interface and used the fdisk utility to partition it. On your system it might look like:

[user@system]$ fdisk /dev/sdb0
Command (m for help): d
  Selected partition 1
  Command (m for help): n
  Command action
     e   extended
     p   primary partition (1-4)
  p
  Partition number (1-4): 1
  First cylinder (1-983, default 1):
  Using default value 1
  Last cylinder or +size or +sizeM or +sizeK (1-983, default 983): +8M
  Command (m for help): n
  Command action
     e   extended
     p   primary partition (1-4)
  p
  Partition number (1-4): 2
  First cylinder (18-983, default 18):
  Using default value 18
  Last cylinder or +size or +sizeM or +sizeK (18-983, default 983):
  Using default value 983
  Command (m for help): w
  The partition table has been altered!
  Calling ioctl() to re-read partition table.
  Syncing disks.

When we got over this the /dev/sdb was partitioned into /dev/sdb0p1 and /dev/sdb0p2. Now we needed to firmat it accordingly. The partition which will hold the kernel needs to be in vfat due to some limitations that can be figured out from the wiki by the curious and the partiotion holding the rootfs needs to be in ext2/ext3.

mkfs.vfat /dev/sdb0p1
mkfs.ext3 /dev/sdb0p2

Now to copy the relevant stuff onto these we need to mount them and perform the copying or writes.

mount /dev/sdb0p2 /mnt/moko
tar -C /mnt/moko/ -xzvf gta02-custom.rootfs.tar.gz OR copy contents of rootfs to the partition.

mount /dev/sdb0p2 /mnt/mokoKernel
cp uImage-gta02-custom.bin /mnt/mokoKernel/uImage.bin

umount /mnt/moko
umount /mnt/mokokernel
sync

UBoot Menu Entries

We need to boot a kernel and provide a root filesystem to it at boot time. UBoot functions as a boot-loader similar to GRUB. It loads the kernel into memory and then passes a list of parameters known as kernel command string to the kernel. Path to root filesystem is a common command string. If you have experienced boot entries in GRUB then we have similar (almost) entries in uboot.

Uboot menu entries are defined by environment variables named menu_X, where X is a number. The value of this variable is a string, <label>:<commands>, where label is the text shown on the screen and commands is a sequence of u-boot commands (delimited by ‘;’ characters) to be executed when the menu item is selected.

Note: When entering a string of commands, the ‘;’ and ‘$’ characters must be backslash-escaped (”\;” and “\$”).

Add a uboot entry

By default both NOR and NAND uboots have an entry for kernel on vfat and rootfs on ext2. If you have this combination then you don’t need to read the following mess. Otherwise here is what you need to do.

To get to the boot promt hold the AUX and power button. Select “Set console to USB” to enter into the boot prompt. The other way to get to the boot prompt is through serial console via Debug Board but we are afraid that we have not yet experimented with that due to pushing deadlines :o)

Now that we are inside the uboot and have the prompt the first thing to do is to set the timeout environment variable because no matter what you are doing the uboot will continue booting after default of 60 seconds.

setenv boot_menu_timeout 99999

This should be enough for us, I guess :o)

To see what the boot environment has

printenv

We had vfat for kernel and ext3 for rootfs so this is what our entry looked like

setenv menu_9 Boot from microSD (FAT+ext3): setenv bootargs \${bootargs_base} rootfstype=ext3 root=/dev/mmcblk0p2 rootdelay=5 \${mtdparts} ro\; mmcinit\; fatload mmc 1 0x32000000 \${sd_image_name}\; bootm 0x32000000

Enter “printenv” to verify and then “saveenv” to save your changes.

Forgot the restart or power off command! Then next time you get to your NAND uboot select the entry that you made earlier and thank us for this tutorial because it takes ours for a newbie to get relevant stuff from the wiki.

Broadband over Power Lines

I came across this technology a few years back. It seemed amazing then and it seems really good now. I think someone needs to consider this for our local community. In my opinion, Amin should take a look at it in detail and from a technical and feasibility perspective since he’s now in the position to talk to businesses and individuals who make the decisions to bring new technology into the country.

How Broadband Over Powerlines Works

With broadband over power lines, or BPL, you can plug your computer into any electrical outlet in your home and instantly have access to high-speed Internet. By combining the technological principles of radio, wireless networking, and modems, developers have created a way to send data over power lines and into homes at speeds between 500 kilobits and 3 megabits per second (equivalent to DSL and cable).

…

By providing high-speed data transmission between all of the electrical plugs in a house, there is the potential to network all kinds of common appliances in a household. If your alarm clock, light switch and coffee maker could talk each other via a high-speed connection, mornings might look a lot different.

[from Howstuffworks.com]

Tpm_dev

Since shaz won’t post on the blog himself, I’ll post it here for him. This comes from his mail to SERG Archive:

This one is specially for Dr.M and Mr.T. The tpm_dev is a kernel module for backward compatibility, when application space made a call to tpm driver i.e. before TIS and 1.2 specification. Do you agree recluze? Now this kernel module is not required when a user space library, “tddl”, can forward userspace requests to “tpmd” the tpm daemon. Do we really need to port tpm_dev module to kernel. It will not be tough but the lesser we handle at a time the better possibility to troubleshoot.

Waiting for ur suggestions. I know its not very difficult to decide but wanted a group decision so that we are even.

IEEE — Standards Organization?!

I don’t mean to be rude but IEEE has done it this time. I was wondering why their publication styles are so typographically ugly but at least that could be justified. Now, they’ve come up with a “new” computer.org home page and … well, just take a look at it:

http://www2.computer.org/portal/web/guest

That’s what bachelor’s students design when they’re learning FrontPage! Sheesh!

Updates?

A.A. everyone,

Are there any updates on the activities of the group? My updates are that I’ve been working on checking the papers and writing the thesis guide for City University. I’m finished with both and should be getting back to Isabelle on Monday, Inshaallah.

Any updates on the projects etc? And what about MMA’s phd defence/final exam?

Meeting this Sunday

Dear Group Members,

I propose a group meeting on this Sunday. The agenda is the coming projects.

Please comment!

MM Alam

4th International Conference on Innovations in Information Technology–Review

Submission

Innovations’2007 seeks original full-length (5 pages maximum) paper proposals describing research in all areas of Information Technology that contribute to the conference theme. All papers will be peer reviewed. Acceptance will be passed on quality, relevance and originality. Accepted papers will be published by IEEE and fully indexed in IEEEXplore. Proposals for tutorials and special sessions are also welcome. The submission guidelines can be found on the conference web site:

http://www.it-innovations.ae

Conference Journals

Selected papers of the conference will be considered for the following International Journals:

- Journal of Communications, Academics Publishers,

- International Journal of Web Services Practices, Computer Science Press,

- International Journal of Information Technology and Web Engineering,

Idea Group Publishing
Important Dates

Papers submissions deadline extended: August 28, 2007

Notification of acceptance: September 22, 2007

Final camera ready papers due: October 27, 2007

I would like all members to give any comments regarding this conference….

amin.